/*
- * Copyright (C) 1995-1998, Index Data
+ * Copyright (C) 1995-1999, Index Data
* All rights reserved.
* Sebastian Hammer, Adam Dickmeiss
*
* $Log: zrpn.c,v $
- * Revision 1.88 1998-10-18 07:54:52 adam
+ * Revision 1.97 1999-10-14 14:33:50 adam
+ * Added truncation 5=106.
+ *
+ * Revision 1.96 1999/09/23 10:05:05 adam
+ * Implemented structure=105 searching.
+ *
+ * Revision 1.95 1999/09/07 07:19:21 adam
+ * Work on character mapping. Implemented replace rules.
+ *
+ * Revision 1.94 1999/07/20 13:59:18 adam
+ * Fixed bug that occurred when phrases had 0 hits.
+ *
+ * Revision 1.93 1999/06/17 14:38:40 adam
+ * Bug fix: Scan SEGV'ed when getting unknown use attribute.
+ *
+ * Revision 1.92 1999/05/26 07:49:13 adam
+ * C++ compilation.
+ *
+ * Revision 1.91 1999/02/02 14:51:13 adam
+ * Updated WIN32 code specific sections. Changed header.
+ *
+ * Revision 1.90 1998/11/16 16:03:43 adam
+ * Moved loggin utilities to Yaz. Was implemented in file zlogs.c.
+ *
+ * Revision 1.89 1998/11/16 10:11:55 adam
+ * Added addtional info for error 114 - unsupported use attribute.
+ *
+ * Revision 1.88 1998/10/18 07:54:52 adam
* Additional info added for diagnostics 114 (Unsupported use attribute) and
* 121 (Unsupported attribute set).
*
*/
#include <stdio.h>
#include <assert.h>
-#ifdef WINDOWS
+#ifdef WIN32
#include <io.h>
#else
#include <unistd.h>
static const char **rpn_char_map_handler (void *vp, const char **from, int len)
{
- struct rpn_char_map_info *p = vp;
+ struct rpn_char_map_info *p = (struct rpn_char_map_info *) vp;
return zebra_maps_input (p->zm, p->reg_type, from, len);
}
while (*src)
{
const char *cp = zebra_maps_output (zh->zebra_maps, reg_type, &src);
- while (*cp)
- *dst++ = *cp++;
+ if (!cp)
+ *dst++ = *src++;
+ else
+ while (*cp)
+ *dst++ = *cp++;
}
*dst = '\0';
}
int *new_term_no;
#endif
p->isam_p_size = 2*p->isam_p_size + 100;
- new_isam_p_buf = xmalloc (sizeof(*new_isam_p_buf) *
- p->isam_p_size);
+ new_isam_p_buf = (ISAM_P *) xmalloc (sizeof(*new_isam_p_buf) *
+ p->isam_p_size);
if (p->isam_p_buf)
{
memcpy (new_isam_p_buf, p->isam_p_buf,
p->isam_p_buf = new_isam_p_buf;
#ifdef TERM_COUNT
- new_term_no = xmalloc (sizeof(*new_term_no) *
- p->isam_p_size);
+ new_term_no = (int *) xmalloc (sizeof(*new_term_no) *
+ p->isam_p_size);
if (p->term_no)
{
memcpy (new_term_no, p->isam_p_buf,
static int grep_handle (char *name, const char *info, void *p)
{
- add_isam_p (name, info, p);
+ add_isam_p (name, info, (struct grep_info *) p);
return 0;
}
return *s0;
}
+/* term_100: handle term, where trunc=none (no operators at all) */
static int term_100 (ZebraMaps zebra_maps, int reg_type,
const char **src, char *dst, int space_split,
char *dst_term)
return i;
}
+/* term_101: handle term, where trunc=Process # */
static int term_101 (ZebraMaps zebra_maps, int reg_type,
const char **src, char *dst, int space_split,
char *dst_term)
return i;
}
-
+/* term_103: handle term, where trunc=re-2 (regular expressions) */
static int term_103 (ZebraMaps zebra_maps, int reg_type, const char **src,
char *dst, int *errors, int space_split,
char *dst_term)
return i;
}
+/* term_103: handle term, where trunc=re-1 (regular expressions) */
static int term_102 (ZebraMaps zebra_maps, int reg_type, const char **src,
char *dst, int space_split, char *dst_term)
{
dst_term);
}
+
+/* term_104: handle term, where trunc=Process # and ! */
+static int term_104 (ZebraMaps zebra_maps, int reg_type,
+ const char **src, char *dst, int space_split,
+ char *dst_term)
+{
+ const char *s0, *s1;
+ const char **map;
+ int i = 0;
+ int j = 0;
+
+ if (!term_pre (zebra_maps, reg_type, src, "#!", "#!"))
+ return 0;
+ s0 = *src;
+ while (*s0)
+ {
+ if (*s0 == '#')
+ {
+ dst[i++] = '.';
+ dst[i++] = '*';
+ dst_term[j++] = *s0++;
+ }
+ else if (*s0 == '!')
+ {
+ dst[i++] = '.';
+ dst_term[j++] = *s0++;
+ }
+ {
+ s1 = s0;
+ map = zebra_maps_input (zebra_maps, reg_type, &s0, strlen(s0));
+ if (space_split && **map == *CHR_SPACE)
+ break;
+ while (s1 < s0)
+ {
+ if (!isalnum (*s1))
+ dst[i++] = '\\';
+ dst_term[j++] = *s1;
+ dst[i++] = *s1++;
+ }
+ }
+ }
+ dst[i] = '\0';
+ dst_term[j++] = '\0';
+ *src = s0;
+ return i;
+}
+
+/* term_105/106: handle term, where trunc=Process # and ! and right trunc */
+static int term_105 (ZebraMaps zebra_maps, int reg_type,
+ const char **src, char *dst, int space_split,
+ char *dst_term, int right_truncate)
+{
+ const char *s0, *s1;
+ const char **map;
+ int i = 0;
+ int j = 0;
+
+ if (!term_pre (zebra_maps, reg_type, src, "*!", "*!"))
+ return 0;
+ s0 = *src;
+ while (*s0)
+ {
+ if (*s0 == '*')
+ {
+ dst[i++] = '.';
+ dst[i++] = '+';
+ dst_term[j++] = *s0++;
+ }
+ else if (*s0 == '!')
+ {
+ dst[i++] = '.';
+ dst_term[j++] = *s0++;
+ }
+ {
+ s1 = s0;
+ map = zebra_maps_input (zebra_maps, reg_type, &s0, strlen(s0));
+ if (space_split && **map == *CHR_SPACE)
+ break;
+ while (s1 < s0)
+ {
+ if (!isalnum (*s1))
+ dst[i++] = '\\';
+ dst_term[j++] = *s1;
+ dst[i++] = *s1++;
+ }
+ }
+ }
+ if (right_truncate)
+ {
+ dst[i++] = '.';
+ dst[i++] = '*';
+ }
+ dst[i] = '\0';
+
+ dst_term[j++] = '\0';
+ *src = s0;
+ return i;
+}
+
+
/* gen_regular_rel - generate regular expression from relation
* val: border value (inclusive)
* islt: 1 if <=; 0 if >=.
dst[dst_p] = '\0';
if (islt)
{
- for (i=1; i<pos; i++)
- strcat (dst, "[0-9]?");
+ /* match everything less than 10^(pos-1) */
+ strcat (dst, "0*");
+ for (i=1; i<pos; i++)
+ strcat (dst, "[0-9]?");
}
else
{
+ /* match everything greater than 10^pos */
for (i = 0; i <= pos; i++)
strcat (dst, "[0-9]");
strcat (dst, "[0-9]*");
}
if (!prefix_len)
{
- zh->errCode = 114;
+ char val_str[32];
+ sprintf (val_str, "%d", use_value);
+ zh->errCode = 114;
+ zh->errString = nmem_strdup (stream, val_str);
return -1;
}
term_dict[prefix_len++] = ')';
logf (LOG_WARN, "dict_lookup_grep err, trunc=eregular: %d",
r);
break;
+ case 104: /* process # and ! in term */
+ term_dict[j++] = '(';
+ if (!term_104 (zh->zebra_maps, reg_type,
+ &termp, term_dict + j, space_split, term_dst))
+ return 0;
+ strcat (term_dict, ")");
+ r = dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ &max_pos, 0, grep_handle);
+ if (r)
+ logf (LOG_WARN, "dict_lookup_grep err, trunc=#/!: %d", r);
+ break;
+ case 105: /* process * and ! in term */
+ term_dict[j++] = '(';
+ if (!term_105 (zh->zebra_maps, reg_type,
+ &termp, term_dict + j, space_split, term_dst, 1))
+ return 0;
+ strcat (term_dict, ")");
+ r = dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ &max_pos, 0, grep_handle);
+ if (r)
+ logf (LOG_WARN, "dict_lookup_grep err, trunc=*/!: %d", r);
+ break;
+ case 106: /* process * and ! in term */
+ term_dict[j++] = '(';
+ if (!term_105 (zh->zebra_maps, reg_type,
+ &termp, term_dict + j, space_split, term_dst, 0))
+ return 0;
+ strcat (term_dict, ")");
+ r = dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ &max_pos, 0, grep_handle);
+ if (r)
+ logf (LOG_WARN, "dict_lookup_grep err, trunc=*/!: %d", r);
+ break;
}
}
*term_sub = termp;
int term_index;
const char *flags = NULL;
- rsfd = xmalloc (sizeof(*rsfd)*rset_no);
- more = xmalloc (sizeof(*more)*rset_no);
- buf = xmalloc (sizeof(*buf)*rset_no);
+ rsfd = (RSFD *) xmalloc (sizeof(*rsfd)*rset_no);
+ more = (int *) xmalloc (sizeof(*more)*rset_no);
+ buf = (struct it_key **) xmalloc (sizeof(*buf)*rset_no);
+ *prox_term = '\0';
for (i = 0; i<rset_no; i++)
{
int j;
- buf[i] = xmalloc (sizeof(**buf));
- rsfd[i] = rset_open (rset[i], RSETF_READ);
- if (!(more[i] = rset_read (rset[i], rsfd[i], buf[i], &term_index)))
- break;
for (j = 0; j<rset[i]->no_rset_terms; j++)
{
const char *nflags = rset[i]->rset_terms[j]->flags;
char *term = rset[i]->rset_terms[j]->name;
int lterm = strlen(term);
- if (length_prox_term)
- prox_term[length_prox_term++] = ' ';
- strcpy (prox_term + length_prox_term, term);
- length_prox_term += lterm;
+ if (lterm + length_prox_term < sizeof(prox_term)-1)
+ {
+ if (length_prox_term)
+ prox_term[length_prox_term++] = ' ';
+ strcpy (prox_term + length_prox_term, term);
+ length_prox_term += lterm;
+ }
if (min_nn > rset[i]->rset_terms[j]->nn)
min_nn = rset[i]->rset_terms[j]->nn;
flags = nflags;
}
}
+ for (i = 0; i<rset_no; i++)
+ {
+ buf[i] = (struct it_key *) xmalloc (sizeof(**buf));
+ rsfd[i] = rset_open (rset[i], RSETF_READ);
+ if (!(more[i] = rset_read (rset[i], rsfd[i], buf[i], &term_index)))
+ break;
+ }
if (i != rset_no)
{
rset_null_parms parms;
xfree (buf[i]);
--i;
}
- parms.rset_term = rset_term_create (prox_term, -1, flags);
+ parms.rset_term = rset_term_create (prox_term, length_prox_term,
+ flags);
parms.rset_term->nn = 0;
result = rset_create (rset_kind_null, &parms);
}
rset_temp_parms parms;
RSFD rsfd_result;
- parms.rset_term = rset_term_create (prox_term, -1, flags);
+ parms.rset_term = rset_term_create (prox_term, length_prox_term,
+ flags);
parms.rset_term->nn = min_nn;
parms.key_size = sizeof (struct it_key);
parms.temp_path = res_get (zh->res, "setTmpDir");
return result;
}
+
+char *normalize_term(ZebraHandle zh, Z_AttributesPlusTerm *zapt,
+ const char *termz, NMEM stream, unsigned reg_id)
+{
+ WRBUF wrbuf = 0;
+ AttrType truncation;
+ int truncation_value;
+ char *ex_list = 0;
+
+ attr_init (&truncation, zapt, 5);
+ truncation_value = attr_find (&truncation, NULL);
+
+ switch (truncation_value)
+ {
+ default:
+ ex_list = "";
+ break;
+ case 101:
+ ex_list = "#";
+ break;
+ case 102:
+ case 103:
+ ex_list = 0;
+ break;
+ case 104:
+ ex_list = "!#";
+ break;
+ case 105:
+ ex_list = "!*";
+ break;
+ }
+ if (ex_list)
+ wrbuf = zebra_replace(zh->zebra_maps, reg_id, ex_list,
+ termz, strlen(termz));
+ if (!wrbuf)
+ return nmem_strdup(stream, termz);
+ else
+ {
+ char *buf = (char*) nmem_malloc (stream, wrbuf_len(wrbuf)+1);
+ memcpy (buf, wrbuf_buf(wrbuf), wrbuf_len(wrbuf));
+ buf[wrbuf_len(wrbuf)] = '\0';
+ return buf;
+ }
+}
+
static RSET rpn_search_APT_phrase (ZebraHandle zh,
Z_AttributesPlusTerm *zapt,
- const char *termz,
+ const char *termz_org,
oid_value attributeSet,
NMEM stream,
int reg_type, int complete_flag,
int num_bases, char **basenames)
{
char term_dst[IT_MAX_WORD+1];
- const char *termp = termz;
RSET rset[60], result;
int i, r, rset_no = 0;
struct grep_info grep_info;
+ char *termz = normalize_term(zh, zapt, termz_org, stream, reg_type);
+ const char *termp = termz;
#ifdef TERM_COUNT
grep_info.term_no = 0;
grep_info.isam_p_indx, term_dst,
strlen(term_dst), rank_type);
assert (rset[rset_no]);
- if (++rset_no >= sizeof(rset)/sizeof(*rset))
+ if (++rset_no >= (int) (sizeof(rset)/sizeof(*rset)))
break;
}
#ifdef TERM_COUNT
static RSET rpn_search_APT_or_list (ZebraHandle zh,
Z_AttributesPlusTerm *zapt,
- const char *termz,
+ const char *termz_org,
oid_value attributeSet,
NMEM stream,
int reg_type, int complete_flag,
int num_bases, char **basenames)
{
char term_dst[IT_MAX_WORD+1];
- const char *termp = termz;
RSET rset[60], result;
int i, r, rset_no = 0;
struct grep_info grep_info;
-
+ char *termz = normalize_term(zh, zapt, termz_org, stream, reg_type);
+ const char *termp = termz;
#ifdef TERM_COUNT
grep_info.term_no = 0;
#endif
grep_info.isam_p_indx, term_dst,
strlen(term_dst), rank_type);
assert (rset[rset_no]);
- if (++rset_no >= sizeof(rset)/sizeof(*rset))
+ if (++rset_no >= (int) (sizeof(rset)/sizeof(*rset)))
break;
}
#ifdef TERM_COUNT
static RSET rpn_search_APT_and_list (ZebraHandle zh,
Z_AttributesPlusTerm *zapt,
- const char *termz,
+ const char *termz_org,
oid_value attributeSet,
NMEM stream,
int reg_type, int complete_flag,
int num_bases, char **basenames)
{
char term_dst[IT_MAX_WORD+1];
- const char *termp = termz;
RSET rset[60], result;
int i, r, rset_no = 0;
struct grep_info grep_info;
+ char *termz = normalize_term(zh, zapt, termz_org, stream, reg_type);
+ const char *termp = termz;
#ifdef TERM_COUNT
grep_info.term_no = 0;
grep_info.isam_p_indx, term_dst,
strlen(term_dst), rank_type);
assert (rset[rset_no]);
- if (++rset_no >= sizeof(rset)/sizeof(*rset))
+ if (++rset_no >= (int) (sizeof(rset)/sizeof(*rset)))
break;
}
#ifdef TERM_COUNT
grep_info.isam_p_indx, term_dst,
strlen(term_dst), rank_type);
assert (rset[rset_no]);
- if (++rset_no >= sizeof(rset)/sizeof(*rset))
+ if (++rset_no >= (int) (sizeof(rset)/sizeof(*rset)))
break;
}
#ifdef TERM_COUNT
if (!sort_sequence->specs)
{
sort_sequence->num_specs = 10;
- sort_sequence->specs = nmem_malloc (stream, sort_sequence->num_specs *
- sizeof(*sort_sequence->specs));
+ sort_sequence->specs = (Z_SortKeySpec **)
+ nmem_malloc (stream, sort_sequence->num_specs *
+ sizeof(*sort_sequence->specs));
for (i = 0; i<sort_sequence->num_specs; i++)
sort_sequence->specs[i] = 0;
}
if (zapt->term->which != Z_Term_general)
i = 0;
else
- i = atoi_n (zapt->term->u.general->buf, zapt->term->u.general->len);
+ i = atoi_n ((char *) zapt->term->u.general->buf,
+ zapt->term->u.general->len);
if (i >= sort_sequence->num_specs)
i = 0;
if (!oid_ent_to_oid (&oe, oid))
return 0;
- sks = nmem_malloc (stream, sizeof(*sks));
- sks->sortElement = nmem_malloc (stream, sizeof(*sks->sortElement));
+ sks = (Z_SortKeySpec *) nmem_malloc (stream, sizeof(*sks));
+ sks->sortElement = (Z_SortElement *)
+ nmem_malloc (stream, sizeof(*sks->sortElement));
sks->sortElement->which = Z_SortElement_generic;
- sk = sks->sortElement->u.generic = nmem_malloc (stream, sizeof(*sk));
+ sk = sks->sortElement->u.generic = (Z_SortKey *)
+ nmem_malloc (stream, sizeof(*sk));
sk->which = Z_SortKey_sortAttributes;
- sk->u.sortAttributes = nmem_malloc (stream, sizeof(*sk->u.sortAttributes));
+ sk->u.sortAttributes = (Z_SortAttributes *)
+ nmem_malloc (stream, sizeof(*sk->u.sortAttributes));
sk->u.sortAttributes->id = oid;
- sk->u.sortAttributes->list =
+ sk->u.sortAttributes->list = (Z_AttributeList *)
nmem_malloc (stream, sizeof(*sk->u.sortAttributes->list));
sk->u.sortAttributes->list->num_attributes = 1;
- sk->u.sortAttributes->list->attributes =
+ sk->u.sortAttributes->list->attributes = (Z_AttributeElement **)
nmem_malloc (stream, sizeof(*sk->u.sortAttributes->list->attributes));
- ae = *sk->u.sortAttributes->list->attributes =
+ ae = *sk->u.sortAttributes->list->attributes = (Z_AttributeElement *)
nmem_malloc (stream, sizeof(**sk->u.sortAttributes->list->attributes));
ae->attributeSet = 0;
- ae->attributeType = nmem_malloc (stream, sizeof(*ae->attributeType));
+ ae->attributeType = (int *)
+ nmem_malloc (stream, sizeof(*ae->attributeType));
*ae->attributeType = 1;
ae->which = Z_AttributeValue_numeric;
- ae->value.numeric = nmem_malloc (stream, sizeof(*ae->value.numeric));
+ ae->value.numeric = (int *)
+ nmem_malloc (stream, sizeof(*ae->value.numeric));
*ae->value.numeric = use_value;
- sks->sortRelation = nmem_malloc (stream, sizeof(*sks->sortRelation));
+ sks->sortRelation = (int *)
+ nmem_malloc (stream, sizeof(*sks->sortRelation));
if (sort_relation_value == 1)
*sks->sortRelation = Z_SortRelation_ascending;
else if (sort_relation_value == 2)
else
*sks->sortRelation = Z_SortRelation_ascending;
- sks->caseSensitivity = nmem_malloc (stream, sizeof(*sks->caseSensitivity));
+ sks->caseSensitivity = (int *)
+ nmem_malloc (stream, sizeof(*sks->caseSensitivity));
*sks->caseSensitivity = 0;
#ifdef ASN_COMPILED
#ifdef ASN_COMPILED
if (*zop->u.prox->u.known != Z_ProxUnit_word)
{
- char *val = nmem_malloc (stream, 16);
+ char *val = (char *) nmem_malloc (stream, 16);
zh->errCode = 132;
zh->errString = val;
sprintf (val, "%d", *zop->u.prox->u.known);
#else
if (*zop->u.prox->proximityUnitCode != Z_ProxUnit_word)
{
- char *val = nmem_malloc (stream, 16);
+ char *val = (char *) nmem_malloc (stream, 16);
zh->errCode = 132;
zh->errString = val;
sprintf (val, "%d", *zop->u.prox->proximityUnitCode);
zh->errString = NULL;
zh->hits = 0;
- sort_sequence = nmem_malloc (nmem, sizeof(*sort_sequence));
+ sort_sequence = (Z_SortKeySpecList *)
+ nmem_malloc (nmem, sizeof(*sort_sequence));
sort_sequence->num_specs = 10;
- sort_sequence->specs = nmem_malloc (nmem, sort_sequence->num_specs *
- sizeof(*sort_sequence->specs));
+ sort_sequence->specs = (Z_SortKeySpec **)
+ nmem_malloc (nmem, sort_sequence->num_specs *
+ sizeof(*sort_sequence->specs));
for (i = 0; i<sort_sequence->num_specs; i++)
sort_sequence->specs[i] = 0;
static int scan_handle (char *name, const char *info, int pos, void *client)
{
int len_prefix, idx;
- struct scan_info *scan_info = client;
+ struct scan_info *scan_info = (struct scan_info *) client;
len_prefix = strlen(scan_info->prefix);
if (memcmp (name, scan_info->prefix, len_prefix))
return 1;
- if (pos > 0)
- idx = scan_info->after - pos + scan_info->before;
+ if (pos > 0) idx = scan_info->after - pos + scan_info->before;
else
idx = - pos - 1;
- scan_info->list[idx].term = odr_malloc (scan_info->odr,
- strlen(name + len_prefix)+1);
+ scan_info->list[idx].term = (char *)
+ odr_malloc (scan_info->odr, strlen(name + len_prefix)+1);
strcpy (scan_info->list[idx].term, name + len_prefix);
assert (*info == sizeof(ISAM_P));
memcpy (&scan_info->list[idx].isam_p, info+1, sizeof(ISAM_P));
term_untrans (zh, reg_type, term_dst, src);
- *dst = nmem_malloc (stream, strlen(term_dst)+1);
+ *dst = (char *) nmem_malloc (stream, strlen(term_dst)+1);
strcpy (*dst, term_dst);
}
char *rank_type = NULL;
int complete_flag;
int sort_flag;
+ *list = 0;
if (attributeset == VAL_NONE)
attributeset = VAL_BIB1;
- zlog_scan (zapt, attributeset);
logf (LOG_DEBUG, "position = %d, num = %d", pos, num);
attr_init (&use, zapt, 1);
zh->errCode = 114;
else
zh->errCode = 121;
+ *num_entries = 0;
+ return;
}
if (zebraExplain_curDatabase (zh->zei, basenames[base_no]))
{
/* prepare dictionary scanning */
before = pos-1;
after = 1+num-pos;
- scan_info_array = odr_malloc (stream, ord_no * sizeof(*scan_info_array));
+ scan_info_array = (struct scan_info *)
+ odr_malloc (stream, ord_no * sizeof(*scan_info_array));
for (i = 0; i < ord_no; i++)
{
int j, prefix_len = 0;
scan_info->after = after;
scan_info->odr = stream;
- scan_info->list = odr_malloc (stream, (before+after)*
- sizeof(*scan_info->list));
+ scan_info->list = (struct scan_info_entry *)
+ odr_malloc (stream, (before+after) * sizeof(*scan_info->list));
for (j = 0; j<before+after; j++)
scan_info->list[j].term = NULL;
dict_scan (zh->dict, termz, &before_tmp, &after_tmp, scan_info,
scan_handle);
}
- glist = odr_malloc (stream, (before+after)*sizeof(*glist));
+ glist = (ZebraScanEntry *)
+ odr_malloc (stream, (before+after)*sizeof(*glist));
/* consider terms after main term */
for (i = 0; i < ord_no; i++)